I have a dataset which consists of daily observations over a 3 month period for different variables. The data are aggregated so I do not have access to user level data.

The problem is that the number of users is not the same every day. I want to conduct an intervention analysis using the ARIMA model. How could I overcome this issue of unequal daily users? Or is it a limitation I cannot avoid?

Thanks in advance.

  • $\begingroup$ Before I respond, what sort of outcome are you trying to forecast? $\endgroup$ – Brash Equilibrium May 18 '18 at 16:09
  • $\begingroup$ I want to conduct an intervention analysis on a past dataset of a video game. My goal is to predict the values of certain metrics before the introduction of an intervention in the game and then compare them with the actual values. That way I can estimate the effect of the event that happened. $\endgroup$ – George Kousvantelos May 21 '18 at 10:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.